US7913090B2 - Authentication systems and authentication method - Google Patents
Authentication systems and authentication method Download PDFInfo
- Publication number
- US7913090B2 US7913090B2 US11/573,859 US57385905A US7913090B2 US 7913090 B2 US7913090 B2 US 7913090B2 US 57385905 A US57385905 A US 57385905A US 7913090 B2 US7913090 B2 US 7913090B2
- Authority
- US
- United States
- Prior art keywords
- biometric subject
- threshold value
- inspected
- predetermined
- subject
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related, expires
Links
- 238000000034 method Methods 0.000 title claims abstract description 21
- 230000000052 comparative effect Effects 0.000 claims description 41
- 238000000605 extraction Methods 0.000 description 11
- 238000012545 processing Methods 0.000 description 11
- 238000010586 diagram Methods 0.000 description 10
- 210000003462 vein Anatomy 0.000 description 7
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000007781 pre-processing Methods 0.000 description 3
- 238000004891 communication Methods 0.000 description 2
- 239000006185 dispersion Substances 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 101000911772 Homo sapiens Hsc70-interacting protein Proteins 0.000 description 1
- 101000661816 Homo sapiens Suppression of tumorigenicity 18 protein Proteins 0.000 description 1
- 230000003467 diminishing effect Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 108090000237 interleukin-24 Proteins 0.000 description 1
- 230000004044 response Effects 0.000 description 1
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/117—Identification of persons
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
Definitions
- the present invention relates to an authentication system for biometric authentication authenticating a biometric subject based on a characteristic quantity acquired from the related biometric subject and a method of same.
- Such authentication includes for example biometric authentication reading biometric data such as fingerprints and vein patterns from the user and performing the authentication based on correlation values of inspected characteristic data showing characteristic features inherent to the user extracted from the biometric data and previously held comparative characteristic data.
- the probability of erroneously judging another person as the true person that is, a False Acceptance Rate (FAR)
- FAR False Acceptance Rate
- FRR False Rejection Rate
- the method of determination of the threshold value includes a variety of methods such as the method shown in the following Patent Document 1.
- Patent Document 1 Japanese Patent No. 3439359
- the present invention has as its object to solve the problems of the prior art explained above by providing an authentication system and an authentication method enabling authentication with a high reliability in comparison with the prior art.
- Another object of the present invention is to provide an authentication system and an authentication method able to realize a desired false acceptance rate or false rejection rate in accordance with the characteristics etc. of the service for which the authentication is used.
- the authentication system of a first aspect of the invention is an authentication system for comparing a correlation value between inspected characteristic data of biometric data acquired from a biometric subject being inspected and comparative data linked with the predetermined biometric subject and a predetermined threshold value linked with the predetermined biometric subject to authenticate whether or not the biometric subject being inspected is the predetermined biometric subject, including: a storing means for storing a threshold value defined so that a value obtained by dividing an absolute value of a difference between a first mean value of a plurality of first correlation values generated by detecting correlation with the comparative data for a plurality of first characteristic data previously acquired from the predetermined biometric subject and the threshold value by a standard deviation of the plurality of first correlation values and values obtained by dividing an absolute value of a difference between a second mean value of a plurality of second correlation values generated by detecting correlation with the comparative data for a plurality of second characteristic data previously acquired from a biometric subject other than the pre
- the mode of operation of the authentication system of the first aspect of the invention is as follows.
- the authenticating means authenticates whether or not the biometric subject being inspected is the predetermined biometric subject by comparing the inspected characteristic data and the threshold value read out from the storing means.
- An authentication method of a second aspect of the invention is an authentication method comparing a correlation value between inspected characteristic data of a biometric subject acquired from the biometric subject being inspected and comparative data linked with a predetermined biometric subject and a predetermined threshold value linked with the predetermined biometric subject to authenticate whether or not the biometric subject being inspected is the predetermined biometric subject, including: a first step of determining a threshold value so that a value obtained by dividing an absolute value of a difference between a first mean value of a plurality of first correlation values generated by detecting correlation with the comparative data for a plurality of first characteristic data previously acquired from the predetermined biometric subject and the threshold value by a standard deviation of the plurality of first correlation values and values obtained by dividing an absolute value of a difference between a second mean value of a plurality of second correlation values generated by detecting correlation with the comparative data for a plurality of second characteristic data previously acquired from a biometric subject other than the predetermined biometric subject and the threshold value by the plurality of second correlation values
- An authentication system of a third aspect of the invention is an authentication system comparing a correlation value between inspected characteristic data of a biometric subject acquired from the biometric subject being inspected and comparative data linked with a predetermined biometric subject and a predetermined threshold value linked with the predetermined biometric subject to authenticate whether or not the biometric subject being inspected is the predetermined biometric subject, including: an inputting means for inputting a false rejection rate showing a probability of certifying that the biometric subject being inspected who is the predetermined biometric subject is not the predetermined biometric subject; a threshold value determining means for determining the threshold value so as to satisfy the false rejection rate input by the inputting means by assuming that a plurality of correlation values generated by detecting correlation with the comparative data for the plurality of characteristic data previously acquired from the predetermined biometric subject are according to a normal profile; and an authenticating means for authenticating whether or not the biometric subject being inspected is the predetermined biometric subject by comparing the correlation value of the inspected characteristic data and the comparative data and the threshold
- a mode of operation of the authentication system of the third aspect of the invention is as follows.
- the inputting means inputs a false rejection rate showing the probability of certifying that a biometric subject being inspected which is the predetermined biometric subject is not the predetermined biometric subject.
- the threshold value determining means determines the threshold value so as to satisfy the false rejection rate input by the inputting means by assuming that a plurality of correlation values generated by detecting correlation with the comparative data for the plurality of characteristic data previously acquired from the predetermined biometric subject are according to a normal profile.
- the authenticating means authenticates whether or not the biometric subject being inspected is the predetermined biometric subject by comparing the correlation value of the inspected characteristic data and the comparative data and the threshold value determined by the threshold value determining means.
- An authentication method of a fourth aspect of the invention is an authentication method comparing a correlation value between inspected characteristic data of a biometric subject acquired from the biometric subject being inspected and comparative data linked with a predetermined biometric subject and a predetermined threshold value linked with the predetermined biometric subject to authenticate whether or not the biometric subject being inspected is the predetermined biometric subject, including: a first step of inputting a false rejection rate showing the probability of certifying that the biometric subject being inspected which is the predetermined biometric subject is not the predetermined biometric subject; a second step of determining the threshold value so as to satisfy the false rejection rate input in the first step by assuming that a plurality of correlation values generated by detecting correlation with the comparative data for the plurality of characteristic data previously acquired from the predetermined biometric subject are according to a normal profile; and a third step of authenticating whether or not the biometric subject being inspected is the predetermined biometric subject by comparing the correlation value of the inspected characteristic data and the comparative data and the threshold value determined in the second
- An authentication system of a fifth aspect of the invention is an authentication system comparing a correlation value between inspected characteristic data of a biometric subject acquired from the biometric subject being inspected and comparative data linked with a predetermined biometric subject and a predetermined threshold value linked with the predetermined biometric subject to authenticate whether or not the biometric subject being inspected is the predetermined biometric subject, including: an inputting means for inputting a false acceptance rate showing the probability of certifying that the biometric subject being inspected who is not the predetermined biometric subject is the predetermined biometric subject; a threshold value determining means for determining the threshold value so as to satisfy the false acceptance rate input by the inputting means by assuming that a plurality of correlation values generated by detecting correlation with the comparative data for the plurality of characteristic data previously acquired from the predetermined biometric subject are according to a normal profile; and an authenticating means for authenticating whether or not the biometric subject being inspected is the predetermined biometric subject by comparing the correlation value of the inspected characteristic data and the comparative data and the threshold value
- a mode of operation of the authentication system of the fifth aspect of the invention is as follows.
- the inputting means inputs a false acceptance rate showing the probability of certifying that a biometric subject being inspected which is not a predetermined biometric subject is the predetermined biometric subject.
- the threshold value determining means determines the threshold value so as to satisfy the false acceptance rate input by the inputting means by assuming that a plurality of correlation values generated by detecting correlation with the comparative data for the plurality of characteristic data previously acquired from the predetermined biometric subject are according to a normal profile.
- the authenticating means compares a correlation value of the inspected characteristic data and the comparative data and the threshold value determined by the threshold value determining means to authenticate whether or not the biometric subject being inspected is the predetermined biometric subject.
- An authentication method of a sixth aspect of the invention is an authentication method for comparing a correlation value between inspected characteristic data of a biometric subject acquired from the biometric subject being inspected and comparative data linked with a predetermined biometric subject and a predetermined threshold value linked with the predetermined biometric subject to authenticate whether or not the biometric subject being inspected is the predetermined biometric subject, including: a first step of inputting a false acceptance rate showing the probability of certifying that the biometric subject being inspected which is not the predetermined biometric subject is the predetermined biometric subject; a second step of determining the threshold value so as to satisfy the false acceptance rate input in the first step by assuming that a plurality of correlation values generated by detecting correlation with the comparative data for the plurality of characteristic data previously acquired from the predetermined biometric subject are according to a normal profile; and a third step of authenticating whether or not the biometric subject being inspected is the predetermined biometric subject by comparing the correlation values of the inspected characteristic data and the comparative data and the threshold value determined in the
- an authentication system and the authentication method able to perform authentication with a higher reliability in comparison with the conventional ones can be provided.
- authentication systems and the authentication methods able to realize an input false acceptance rate or a false rejection rate can be provided.
- FIG. 1 is a view of the overall configuration of an authentication system of a first embodiment of the present invention.
- FIG. 2 is a flow chart for explaining processing of a threshold value determination unit shown in FIG. 1 .
- FIG. 3 is a flow chart for explaining the processing of step ST 3 shown in FIG. 2 .
- FIG. 4 is a diagram for explaining characteristics of another profile.
- FIG. 5 is a diagram for explaining characteristics of another profile where a skewness is negative.
- FIG. 6 is a diagram for explaining characteristics of a deformed other profile where the skewness is negative.
- FIG. 7 is a diagram for explaining a threshold value when using a deformed other profile where the skewness is negative.
- FIG. 8 is a diagram for explaining characteristics of another profile where the skewness is positive.
- FIG. 9 is a diagram for explaining characteristics of a deformed other profile where the skewness is positive.
- FIG. 10 is a diagram for explaining the threshold value when using a deformed other profile where the skewness is positive.
- FIG. 11 is a view of the overall configuration of an authentication system of a second embodiment of the present invention.
- FIG. 12 is a flow chart for explaining the processing of step ST 3 shown in FIG. 2 in the authentication system shown in FIG. 11 .
- FIG. 13 is a view of the overall configuration of an authentication system of a third embodiment of the present invention.
- FIG. 14 is a diagram for explaining relationships of another profile and a true profile and FRR and FAR.
- FIG. 15 is a diagram for explaining a normal profile table data TABLE stored in a memory shown in FIG. 13 .
- FIG. 16 is a flow chart for explaining pre-processing of the threshold value determination unit of the authentication system shown in FIG. 15 .
- FIG. 17 is a flow chart for explaining the processing of the threshold value determination unit of the authentication systems shown in FIG. 15 .
- FIG. 18 is a diagram for explaining an example of an experiment of the authentication system shown in FIG. 15 .
- a memory 18 corresponds to the storing means of the first aspect of the invention
- an authentication unit 20 corresponds to the authenticating means of the first aspect of the invention
- a threshold value determination unit 22 corresponds to the threshold value determining means of the first aspect of the invention.
- True person sample data Ct 1 to Ctn correspond to the first characteristic data of the first aspect of the invention
- other sample data Co 1 to Com correspond to the second characteristic data of the first aspect of the invention
- correlation data Ft correspond to the first correlation values of the first aspect of the invention
- correlation data Fo correspond to the second correlation values of the first aspect of the invention.
- a mean value ⁇ t corresponds to the first mean value of the first aspect of the invention
- a mean value ⁇ o corresponds to the second mean value of the first aspect of the invention.
- a threshold value Xth corresponds to the threshold value of the first aspect of the invention.
- FIG. 1 is a view of the configuration of an authentication system 1 according to an embodiment of the present invention.
- the authentication system 1 has for example a biometric data reading unit 12 , characteristic extraction unit 14 , correlation value calculation unit 16 , memory 18 , authentication unit 20 , and threshold value determination unit 22 .
- the characteristic extraction unit 14 , correlation value calculation unit 16 , memory 18 , authentication unit 20 , and threshold value determination unit 22 are realized by executing a program by dedicated hardware such as an electronic circuit or processing circuit.
- the biometric data reading unit 12 reads for example the fingerprint or vein pattern or other the biometric data from a human finger or other biometric subject 10 and outputs the related read biometric data S 12 to the characteristic extraction unit 14 .
- the characteristic extraction unit 14 extracts inspected characteristic data S 14 showing a characteristic feature such as a branch point and end point of a fingerprint and vein pattern from the biometric data S 12 input from the biometric data reading unit 12 and outputs this to the correlation value calculation unit 16 .
- the correlation value calculation unit 16 detects correlation data Ft indicating the correlation value of the inspected characteristic data S 14 input from the characteristic extraction unit 14 and the reference characteristic data Cref read out from the memory 18 and outputs this to the authentication unit 20 .
- the memory 18 stores the reference characteristic data Cref and a threshold value Xth written from the threshold value determination unit 22 .
- the authentication unit 20 judges whether or not the correlation value indicated by the correlation data Ft input from the correlation value calculation unit 16 is larger than the threshold value Xth, certifies that the biometric subject 10 is legitimate when judging that the correlation value is larger, and certifies that the biometric subject 10 is not legitimate when not judging so.
- the threshold value determination unit 22 calculates the threshold value Xth as explained below based on the true sample data Ct 1 to Ctn of the characteristic data previously acquired from the biometric subject 10 a plurality of times and the other sample data Co 1 to Com of the characteristic data previously acquired from a biometric subject other than the biometric subject 10 (other person) a plurality of times and writes this into the memory 18 .
- FIG. 2 is a flow chart for explaining the processing of the threshold value determination unit 22 shown in FIG. 1 .
- Step ST 1
- the threshold value determination unit 22 receives as input the true sample data Ct 1 to Ctn from another apparatus via the memory 18 or the network etc.
- Step ST 2
- the threshold value determination unit 22 receives as input the other sample data Co 1 to Com from another apparatus via the memory 18 or the network etc.
- Step ST 3
- the threshold value determination unit 22 calculates the threshold value Xth based on the true sample data Ct 1 to Ctn input at step ST 1 and the other sample data Co 1 to Com input at step ST 2 .
- Step ST 4
- the threshold value determination unit 22 writes (sets) the threshold value Xth calculated at step ST 3 in the memory 18 .
- step ST 3 shown in FIG. 2 will be explained in detail.
- FIG. 3 is a flow chart for explaining step ST 3 shown in FIG. 2 .
- Step ST 11
- the threshold value determination unit 22 calculates correlation data Ft 1 to Ftn indicating correlation values with the reference characteristic data Cref read out from the memory 18 for each of the true sample data Ct 1 to Ctn input at step ST 1 shown in FIG. 2 .
- Step ST 12
- the threshold value determination unit 22 calculates correlation data Fo 1 to Fom indicating correlation values with the reference characteristic data Cref read out from the memory 18 for each of the other sample data Co 1 to Com input at step ST 2 shown in FIG. 2 .
- Step ST 13
- the threshold value determination unit 22 calculates the mean value ⁇ t of n number of correlation data Ft 1 to Ftn calculated at step ST 11 .
- Step ST 14
- the threshold value determination unit 22 calculates the mean value ⁇ o of m number of correlation data Fo 1 to Fom calculated at step ST 12 .
- Step ST 15
- the threshold value determination unit 22 calculates the standard deviation ⁇ t of the correlation data for the true person based on the following equation (1) based on the correlation data Ft 1 to Ftn calculated at step ST 11 and the mean value ⁇ t calculated at step ST 13 .
- Step ST 16
- the threshold value determination unit 22 calculates the standard deviation ⁇ o of the correlation data for the other person based on the following equation (2) based on the correlation data Fo 1 to Fom calculated at step ST 12 and the mean value ⁇ o calculated at step ST 14 .
- Step ST 17
- the threshold value determination unit 22 calculates a value X satisfying the following equation (3) based on the mean value ⁇ t calculated at step ST 13 and the standard deviation ⁇ t calculated at step ST 15 .
- the left side indicates a Mahalanobis distance according to the true profile
- the right side indicates the Mahalanobis distance according to the other profile.
- the threshold value determination unit 22 calculates the value X based on the following equation (7).
- Step ST 18
- the threshold value determination unit 22 modifies the value X calculated at step ST 17 to the threshold value Xth.
- the threshold value determination unit 22 assumes that the true profile defined by the correlation data Ft 1 to Ftn and the other profile defined by the correlation data Fo 1 to Fom are normal profiles and calculates the threshold value Xth based on the mean values ⁇ t and ⁇ o and standard deviations ⁇ t and ⁇ o of these.
- the threshold value determination unit 22 shown in FIG. 1 generates the threshold value Xth as explained by using FIG. 2 and FIG. 3 and writes this into the memory 18 .
- the biometric data reading unit 12 reads biometric data such as a fingerprint or vein pattern from a human finger or other biometric subject 10 and outputs the read biometric data S 12 to the characteristic extraction unit 14 .
- the characteristic extraction unit 14 extracts the inspected characteristic data S 14 indicating a characteristic feature such as a branch point or end point of a fingerprint or vein pattern from the biometric data S 12 input from the biometric data reading unit 12 and outputs this to the correlation value calculation unit 16 .
- the correlation value calculation unit 16 detects the correlation data Ft indicating the correlation value of the inspected characteristic data S 14 input from the characteristic extraction unit 14 and the reference characteristic data Cref read out from the memory 18 and outputs this to the authentication unit 20 .
- the authentication unit 20 judges whether or not the correlation value indicated by the correlation data Ft input from the correlation value calculation unit 16 is larger than the threshold value Xth, certifies that the biometric subject 10 is legitimate when judging that the correlation value is larger, and certifies that the biometric subject 10 is not legitimate when not judging so.
- the threshold value determination unit 22 determines the threshold value Xth (X) so that it indicates the Mahalanobis distance according to the true profile and it coincides with the Mahalanobis distance according to the other profile.
- the false acceptance rate FAR and the false rejection rate FRR can be made to schematically coincide and balanced high precision authentication can be performed.
- the true profile TP corresponds to the first normal profile of the first aspect of the invention
- the other profile OP corresponds to the second normal profile of the first aspect of the invention
- the deformed other profile corresponds to the third normal profile of the first aspect of the invention.
- the threshold value determination unit 22 calculated the mean value ⁇ o and the standard deviation ⁇ o by using the other profile defined according to the correlation data Fo 1 to Fom as it was.
- the other profile is deformed by using the skewness, calculates the mean value ⁇ o and the standard deviation ⁇ o based on the deformed other profile, and stably suppresses the FAR/FRR low.
- the skewness is the value expressing the left and right symmetry of the profile, becomes zero in the case of the left and right symmetry as in the normal profile, becomes the profile biased to the right as shown in for example FIG. 4A in the case of a negative value, and becomes the profile biased to the left as shown in for example FIG. 4B in the case of a positive value, and the spread of the profile becomes the inverse direction thereof.
- the Mahalanobis distance from the center OP_C of the other profile OP is used, but as shown in FIG. 5 , when the other profile OP has a biased profile, it becomes possible to lower the FAR without raising the FRR by utilizing that eccentricity.
- the frequency is sharply lowered on the right side from the profile center of the other profile OP and there is no spread of the profile.
- the frequency gently falls in comparison with the right side and also the spread of profile is large.
- the standard deviation of the other profile OP is calculated by using these left and right data, but in confirming the true property, by using only the data on the true side, that is, the sharply changing data, to calculate the standard deviation again, it becomes possible to obtain the true side other profile data in a conscious form.
- the skewness becomes negative.
- the profile ⁇ c 2 is indicated by the following equation (9).
- a threshold value thr 1 indicates the distance 3 times the standard deviation from the center of the other profile and a threshold value thr 2 indicates the distance 4.27 times the standard deviation from the profile center of the other person.
- the FAR becomes 0.001% when converted from the normal profile table.
- the threshold value thr 2 When using the threshold value thr 2 defined based on the other profile OP, the threshold value thr 2 fully enters into the true profile and the false rejection rate FRR becomes relatively large. On the other hand, when determining the threshold value thr 2 based on the deformed other profile OPA, the false rejection rate FRR can be sufficiently, lowered with almost no rise of the false acceptance rate FAR.
- the profile becomes as in FIG. 9 .
- the center of the other profile OP on the true side is made the position where the degree of the other profile becomes the maximum value.
- the threshold value thr 1 indicates the distance 3 times the standard deviation from the center of the other profile
- the threshold value thr 2 indicates the distance 4.27 times the standard deviation from the profile center of the other person.
- FAR becomes 0.001% when converted from the normal profile table.
- the threshold value thr 2 defined based on the other profile OP When the threshold value thr 2 defined based on the other profile OP is used, the threshold value thr 2 is sufficiently apart from the true profile, and the false rejection rate FRR is sufficiently small.
- the threshold value thr 2 A is determined based on the deformed other profile OPA, both of the false acceptance rate FAR and the false rejection rate FRR rise. That is, when the threshold value is determined by the whole profile, there is a worry of the set FAR and FRR insufficiently functioning. According to the present embodiment, the problem of a threshold value lower than the original threshold value being set and actually becoming an obstacle at the time of authentication can be solved.
- FIG. 11 is a view of the configuration of the authentication system 101 according to the embodiment of the present invention.
- the authentication system 101 has for example a biometric data reading unit 12 , characteristic extraction unit 14 , correlation value calculation unit 16 , memory 18 , authentication unit 20 , and threshold value determination unit 122 .
- FIG. 11 parts given the same notations as those of FIG. 1 are the same as those explained in the first embodiment.
- the authentication system 101 is different in the threshold value determination unit 122 from the threshold value determination unit 22 of the first embodiment.
- the threshold value determination unit 122 is realized by running a program by dedicated hardware such as an electronic circuit or processing circuit.
- threshold value determination unit 122 will be explained in detail.
- the threshold value determination unit 122 calculates the threshold value Xth as will be explained below based on the true sample data Ct 1 to Ctn of the characteristic data previously acquired from a biometric subject 10 a plurality of times and the other sample data Co 1 to Com of the characteristic data previously acquired from a biometric subject other than the biometric subject 10 (other person) a plurality of times and writes this into the memory 18 .
- the threshold value determination unit 122 does not use the other sample data Co 1 to Com as they are, but generates the other profile OPA which becomes linearly symmetric with the maximum degree about the other profile OP defined by the other sample data Co 1 to Com and calculates the threshold value Xth by using this other profile OPA.
- FIG. 12 is a flow chart for explaining the processing of the threshold value determination unit 122 .
- Step ST 32 shown in FIG. 12 is the same as step ST 11 shown in FIG. 3 .
- steps ST 34 to ST 39 shown in FIG. 12 are the same as steps ST 13 to ST 18 shown in FIG. 3 .
- the threshold value determination unit 122 generates the other profile OPA which becomes linearly symmetric about the maximum degree in the other profile OP defined by the other sample data Co 1 to Com input at step ST 2 shown in FIG. 2 as explained above.
- the threshold value determination unit 122 calculates the correlation data Fo 1 to Fom indicating the correlation values with the reference characteristic data Cref read out from the memory 18 for each of the other sample data Co 1 to Com composing the other profile OPA generated at step ST 31 .
- the threshold value determination unit 122 determines the threshold Xth based on the deformed other profile OPA by using the skewness of the other profile OP for the other profile OP, therefore the FAR/FRR can be stably suppressed low.
- the present embodiment corresponds to the third to sixth aspects of the invention.
- An input unit 221 corresponds to the inputting means of the third and fifth embodiments
- an authentication unit 20 corresponds to the authenticating means of the third and fifth aspects of the invention
- a threshold value determination unit 222 corresponds to the threshold value determining means of the third and fifth aspects of the invention.
- a memory 18 corresponds to the storing means of the third and fifth aspects of the invention.
- the false acceptance rate FAR corresponds to the false acceptance rate of the present invention
- the false rejection rate FRR corresponds to the false rejection rate
- FIG. 13 is a view of the configuration of an authentication system 201 according to this embodiment of the present invention.
- the authentication systems 201 has for example a biometric data reading unit 12 , characteristic extraction unit 14 , correlation value calculation unit 16 , memory 18 , authentication unit 20 , input unit 221 , and threshold value determination unit 222 .
- FIG. 13 parts given the same notations as those in FIG. 1 are the same as those explained in the first embodiment.
- the authentication system 201 has an input unit 221 .
- the threshold value determination unit 222 is different from the threshold value determination unit 22 of the first embodiment.
- the threshold value determination unit 222 is realized by executing a program by dedicated hardware such as an electronic circuit or processing circuit.
- the input unit 221 is an inputting means such as a keyboard and mouse and inputs a false acceptance rate or false rejection rate FRR in response to the operation of the user.
- the threshold value determination unit 222 determines the threshold value Xth so as to satisfy the false acceptance rate FAR or the false rejection rate FRR input by the input unit 221 assuming that the true profile TP and the other profile OP are according to the normal profiles.
- the false acceptance rate FAR indicates the ratio with respect to the value obtained by integrating the other profile OP from the threshold value Xth to 1 for the entire other profile OP, that is, the ratio of the area of the right side in the figure from the threshold value Xth of the other profile OP.
- the false rejection rate FRR indicates the ratio with respect to the value obtained by integrating the true profile TP from 0 to the threshold value Xth for the entire true profile TP, that is, the area of the left side in the figure from the threshold value Xth of the true profile TP.
- the threshold value determination unit 222 specifies a value near a value “FRR/100” corresponding to the false rejection rate FRR from the normal profile table data TABLE shown in FIG. 15 and designates the specified value as a Mahalanobis distance dt.
- the normal profile table data TABLE is stored in for example the memory 18 shown in FIG. 13 .
- the threshold value determination unit 222 specifies the false acceptance rate FAR as the value corresponding to the Mahalanobis distance do from the normal profile table data TABLE shown in FIG. 15 .
- the threshold value determination unit 222 specifies a value near a value “FAR/100” corresponding to the false acceptance rate FAR from the normal profile table data TABLE shown in FIG. 15 and determines the specified value as the Mahalanobis distance do.
- the threshold value determination unit 222 specifies the false rejection rate FRR as the value corresponding to the Mahalanobis distance dt from the normal profile table data TABLE shown in FIG. 15 .
- the Mahalanobis distance may be calculated by approximation based on the input false acceptance rate FAR or the false rejection rate FRR.
- FIG. 16 is a flow chart for explaining the pre-processing of the threshold value determination unit 222 shown in FIG. 13 .
- the threshold value determination unit 222 performs the pre-processing shown in FIG. 16 before the false acceptance rate FAR or the false rejection rate FRR is input.
- Steps ST 51 to ST 56 shown in FIG. 16 are the same as steps ST 11 to ST 16 explained in the first embodiment by using FIG. 3 .
- FIG. 17 is a flow chart for explaining the processing of the threshold value determination unit 222 when the false acceptance rate FAR or the false rejection rate FRR is input.
- Step ST 61
- the threshold value determination unit 222 judges whether or not the input unit 221 inputted the false rejection rate FRR, proceeds to step ST 62 when judging it did, and proceeds to step ST 64 when judging it did not.
- Step ST 62
- the threshold value determination unit 222 specifies the value near “FRR/100” from the normal profile table data TABLE stored in the memory 18 based on the input false rejection rate FRR and designates the specified value as the Mahalanobis distance dt.
- Step ST 63
- the threshold value determination unit 222 performs the computation according to above equation (11) by using the Mahalanobis distance dt acquired at step ST 62 , the standard deviation ⁇ t calculated at step ST 55 shown in FIG. 16 , and the mean value ⁇ t calculated at step ST 53 to calculate the threshold value Xth.
- the threshold value determination unit 222 calculates the false acceptance rate FAR by using the threshold value Xth as previously explained and changes the FRR and newly calculates the threshold value Xth where this does not satisfy the predetermined condition.
- Step ST 64
- the threshold value determination unit 222 judges whether or not the input unit 221 inputted the false acceptance rate FAR, proceeds to step ST 65 when judging it did, and returns to step ST 61 when judging it did not.
- Step ST 65
- the threshold value determination unit 222 specifies the value near “FAR/100” from the normal profile table data TABLE stored in the memory 18 based on the input false acceptance rate FAR and designates the specified value as the Mahalanobis distance do.
- Step ST 66
- the threshold value determination unit 222 performs the computation according to equation (13) by using the Mahalanobis distance ot acquired at step ST 65 , the standard deviation ⁇ o calculated at step ST 56 shown in FIG. 16 , and the mean value ⁇ o calculated at step ST 54 to calculate the threshold value Xth.
- the threshold value determination unit 222 calculates the false acceptance rate FAR by using the threshold value Xth as previously explained and changes the FRR and newly calculates the threshold value Xth where this does not satisfy the predetermined condition.
- the threshold value Xth can be set so as to realize the false acceptance rate FAR or false rejection rate FRR input via the input unit 221 .
- the false acceptance rate FAR may be high, but the false rejection rate FRR is desired to be lowered.
- the false rejection rate FRR may be high, but the false acceptance rate FAR is desired to be lowered. Authentication tailored to these is possible.
- FIG. 18 An image emphasizing only the finger vein pattern from a finger vein pattern image was used as the characteristic quantity of the true person.
- the correlation values of the images of the characteristic quantities were used to differentiate the true person and another person.
- FIG. 18 several sets of true data are collected from seven subjects (A to G). There are exactly the number of true reference data of the true data. Points concentrated on the left side represent the correlation with the other person data, and points dispersed on the right side mean the correlation with the true data.
- a middle of a bar at the other person data means an average of correlation values with the other person data and means distances of ⁇ and 3 ⁇ from the mean value.
- the true person and another person can be reliably separated with the same threshold value for any subject. That is, they may be separated when setting the threshold value at approximately 0.43. This value is determined by viewing the lowest value of the correlation values of the subject E and the true person. When the threshold value is further lowered, it becomes the maximum value of the correlation values of the subject C with the other person. That is, when the threshold value is lowered in order to avoid false rejection of the subject E, false acceptance of the subject C will be permitted. In this way, irrespective of the characteristic values extracted with the same measure, a variation occurs according to subjects, and it may be difficult to unambiguously determine the threshold value.
- the maximum value of the correlation values with the other person is high, but the minimum value of the correlation values with the true person is large. That is, limited to the subject C, differentiating between the true person and another person becomes easy even when the threshold value is larger than 0.43 explained before. Further, limited to the subject D, the mean value is a little high, but the dispersion is small, therefore the threshold value can be suppressed low.
- the present invention is not limited to the above embodiments.
- the threshold value Xth was determined in the threshold value determination units 22 , 122 , and 222 in the authentication systems 1 , 101 , and 201 was exemplified, but the threshold value determination units 22 , 122 , and 222 may be assembled in apparatuses other than the authentication systems 1 , 101 , and 201 , for example service apparatuses with which the authentication systems 1 , 101 , and 201 communicate, and the authentication systems 1 , 101 , and 201 may receive as input the threshold value Xth from the related apparatuses.
- the present invention can be applied to a system for authentication based on biometric data.
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Multimedia (AREA)
- Human Computer Interaction (AREA)
- Probability & Statistics with Applications (AREA)
- Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Animal Behavior & Ethology (AREA)
- Biomedical Technology (AREA)
- Heart & Thoracic Surgery (AREA)
- Medical Informatics (AREA)
- Biophysics (AREA)
- Surgery (AREA)
- Pathology (AREA)
- General Health & Medical Sciences (AREA)
- Public Health (AREA)
- Veterinary Medicine (AREA)
- Collating Specific Patterns (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
- Image Analysis (AREA)
Abstract
Description
[Equation 4]
μt>μO (4)
[Equation 6]
σo(μt−X)=σt(X−μo) (6)
[Equation 11]
Xth=μt−σtdt (11)
[Equation 13]
Xth=μo+σodo (13)
Claims (11)
Applications Claiming Priority (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2004239439A JP4140582B2 (en) | 2004-08-19 | 2004-08-19 | Authentication apparatus and authentication method |
JP2004-239439 | 2004-08-19 | ||
PCT/JP2005/014816 WO2006019045A1 (en) | 2004-08-19 | 2005-08-12 | Authentication device and authentication method |
Publications (2)
Publication Number | Publication Date |
---|---|
US20080040614A1 US20080040614A1 (en) | 2008-02-14 |
US7913090B2 true US7913090B2 (en) | 2011-03-22 |
Family
ID=35907433
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US11/573,859 Expired - Fee Related US7913090B2 (en) | 2004-08-19 | 2005-08-12 | Authentication systems and authentication method |
Country Status (8)
Country | Link |
---|---|
US (1) | US7913090B2 (en) |
EP (1) | EP1785939A4 (en) |
JP (1) | JP4140582B2 (en) |
KR (1) | KR101101024B1 (en) |
CN (1) | CN101006466B (en) |
HK (1) | HK1102149A1 (en) |
TW (1) | TW200620950A (en) |
WO (1) | WO2006019045A1 (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20080295152A1 (en) * | 2007-05-25 | 2008-11-27 | Laurel Precision Machines Co., Ltd. | Safety management system |
US20090027276A1 (en) * | 2007-07-23 | 2009-01-29 | Ming-Tsan Tseng | Linear polarized patch antenna structure |
US8297399B2 (en) | 2007-05-23 | 2012-10-30 | Laurel Precision Machines Co., Ltd. | Safety management system |
Families Citing this family (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7206448B2 (en) | 2002-02-28 | 2007-04-17 | At&T Corp. | System and method for using pattern vectors for video and image coding and decoding |
US7627144B2 (en) * | 2004-08-19 | 2009-12-01 | Sony Corporation | Method and apparatus for performing identity authentication |
CN101187985B (en) * | 2006-11-17 | 2012-02-01 | 东软集团股份有限公司 | Method and device for classification boundary of identifying object classifier |
WO2008143270A1 (en) * | 2007-05-15 | 2008-11-27 | Sony Corporation | Registering device, checking device, program, and data structure |
KR101527711B1 (en) * | 2007-12-20 | 2015-06-11 | 코닌클리케 필립스 엔.브이. | Defining classification thresholds in template protection systems |
JP2009163555A (en) * | 2008-01-08 | 2009-07-23 | Omron Corp | Face collation apparatus |
JP2010092122A (en) * | 2008-10-03 | 2010-04-22 | Fujitsu Ltd | Authentication system, biometrics authentication device, and biometrics authentication method |
JP5391631B2 (en) | 2008-10-03 | 2014-01-15 | 富士通株式会社 | Parameter control device, parameter control program, and multistage collation device |
JP2012018432A (en) | 2010-07-06 | 2012-01-26 | Hitachi Omron Terminal Solutions Corp | Biological authentication device and biological authentication method |
AU2012283743B2 (en) * | 2011-07-08 | 2015-02-26 | Australian Nuclear Science And Technology Organisation | Radionuclide detection and identification |
KR101736710B1 (en) * | 2015-08-07 | 2017-05-17 | 주식회사 슈프리마 | Method and apparatus for management of biometric data |
JP6707271B2 (en) * | 2015-10-20 | 2020-06-10 | 株式会社ステップワン | Personal authentication device, personal authentication method, and personal authentication program |
CN105404807B (en) * | 2015-12-08 | 2019-02-05 | Oppo广东移动通信有限公司 | Promote the method, device and mobile terminal of fingerprint recognition performance |
JP7179663B2 (en) * | 2019-03-28 | 2022-11-29 | 三菱重工業株式会社 | PLANT MONITORING DEVICE, PLANT MONITORING METHOD, AND PROGRAM |
Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US4407299A (en) | 1981-05-15 | 1983-10-04 | The Children's Medical Center Corporation | Brain electrical activity mapping |
US5941820A (en) | 1994-01-03 | 1999-08-24 | Zimmerman; Steven | Medical data display method |
US6028949A (en) * | 1997-12-02 | 2000-02-22 | Mckendall; Raymond A. | Method of verifying the presence of an eye in a close-up image |
US20020118864A1 (en) | 2001-02-28 | 2002-08-29 | Kenji Kondo | Personal authentication method and device |
JP2003044832A (en) | 2001-07-31 | 2003-02-14 | Kobe Steel Ltd | Method, device and program for setting inspection parameter |
JP2003067340A (en) | 2001-08-28 | 2003-03-07 | Mitsubishi Electric Corp | Selection system for authentication, and authentication system |
JP2003099780A (en) | 2001-09-21 | 2003-04-04 | Nippon Signal Co Ltd:The | Access control system |
JP3439359B2 (en) | 1998-12-18 | 2003-08-25 | 日本電気株式会社 | Personal identification method, personal identification device, and recording medium |
US20040034479A1 (en) | 2002-08-07 | 2004-02-19 | Akihiro Shimase | Sample dispensing apparatus and automatic analyzer using the same |
US7627144B2 (en) | 2004-08-19 | 2009-12-01 | Sony Corporation | Method and apparatus for performing identity authentication |
Family Cites Families (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5917928A (en) * | 1997-07-14 | 1999-06-29 | Bes Systems, Inc. | System and method for automatically verifying identity of a subject |
JP3620391B2 (en) * | 2000-02-23 | 2005-02-16 | 日本電気株式会社 | Fingerprint input device, image determination method used therefor, and recording medium recording control program thereof |
CN1442823A (en) * | 2002-12-30 | 2003-09-17 | 潘国平 | Individual identity automatic identification system based on iris analysis |
-
2004
- 2004-08-19 JP JP2004239439A patent/JP4140582B2/en not_active Expired - Fee Related
-
2005
- 2005-08-12 EP EP05780256A patent/EP1785939A4/en not_active Withdrawn
- 2005-08-12 CN CN2005800282543A patent/CN101006466B/en not_active Expired - Fee Related
- 2005-08-12 KR KR1020077003803A patent/KR101101024B1/en active IP Right Grant
- 2005-08-12 WO PCT/JP2005/014816 patent/WO2006019045A1/en active Application Filing
- 2005-08-12 US US11/573,859 patent/US7913090B2/en not_active Expired - Fee Related
- 2005-08-17 TW TW094127975A patent/TW200620950A/en not_active IP Right Cessation
-
2007
- 2007-09-19 HK HK07110173.1A patent/HK1102149A1/en not_active IP Right Cessation
Patent Citations (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US4407299A (en) | 1981-05-15 | 1983-10-04 | The Children's Medical Center Corporation | Brain electrical activity mapping |
US5941820A (en) | 1994-01-03 | 1999-08-24 | Zimmerman; Steven | Medical data display method |
US6028949A (en) * | 1997-12-02 | 2000-02-22 | Mckendall; Raymond A. | Method of verifying the presence of an eye in a close-up image |
JP3439359B2 (en) | 1998-12-18 | 2003-08-25 | 日本電気株式会社 | Personal identification method, personal identification device, and recording medium |
US20020118864A1 (en) | 2001-02-28 | 2002-08-29 | Kenji Kondo | Personal authentication method and device |
JP2002259981A (en) | 2001-02-28 | 2002-09-13 | Matsushita Electric Ind Co Ltd | Individual authentication method and device |
JP2003044832A (en) | 2001-07-31 | 2003-02-14 | Kobe Steel Ltd | Method, device and program for setting inspection parameter |
JP2003067340A (en) | 2001-08-28 | 2003-03-07 | Mitsubishi Electric Corp | Selection system for authentication, and authentication system |
US20030051147A1 (en) | 2001-08-28 | 2003-03-13 | Mitsubishi Denki Kabushiki Kaisha | Authentication-selection system, and authentication system |
JP2003099780A (en) | 2001-09-21 | 2003-04-04 | Nippon Signal Co Ltd:The | Access control system |
US20040034479A1 (en) | 2002-08-07 | 2004-02-19 | Akihiro Shimase | Sample dispensing apparatus and automatic analyzer using the same |
US7627144B2 (en) | 2004-08-19 | 2009-12-01 | Sony Corporation | Method and apparatus for performing identity authentication |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8297399B2 (en) | 2007-05-23 | 2012-10-30 | Laurel Precision Machines Co., Ltd. | Safety management system |
US20080295152A1 (en) * | 2007-05-25 | 2008-11-27 | Laurel Precision Machines Co., Ltd. | Safety management system |
US8799669B2 (en) * | 2007-05-25 | 2014-08-05 | Laurel Precision Machines Co., Ltd. | Safety management system |
US20090027276A1 (en) * | 2007-07-23 | 2009-01-29 | Ming-Tsan Tseng | Linear polarized patch antenna structure |
Also Published As
Publication number | Publication date |
---|---|
TWI309943B (en) | 2009-05-11 |
US20080040614A1 (en) | 2008-02-14 |
TW200620950A (en) | 2006-06-16 |
KR101101024B1 (en) | 2011-12-29 |
HK1102149A1 (en) | 2007-11-09 |
CN101006466A (en) | 2007-07-25 |
EP1785939A4 (en) | 2012-07-04 |
JP2006059071A (en) | 2006-03-02 |
JP4140582B2 (en) | 2008-08-27 |
CN101006466B (en) | 2010-11-03 |
KR20070040815A (en) | 2007-04-17 |
EP1785939A1 (en) | 2007-05-16 |
WO2006019045A1 (en) | 2006-02-23 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US7913090B2 (en) | Authentication systems and authentication method | |
US7151846B1 (en) | Apparatus and method for matching fingerprint | |
US8816818B2 (en) | Biometric authentication device and method, computer-readable recording medium recorded with biometric authentication computer program, and computer system | |
EP1183638B1 (en) | Method and apparatus for creating a composite fingerprint image | |
US6591224B1 (en) | Biometric score normalizer | |
Martens et al. | On-line signature verification by dynamic time-warping | |
US20060117188A1 (en) | Biometric print quality assurance | |
JP5363587B2 (en) | Biometric information registration method, biometric authentication method, and biometric authentication device | |
US11062120B2 (en) | High speed reference point independent database filtering for fingerprint identification | |
EP1989813A2 (en) | Method and apparatus for rolling enrollment for signature verification | |
EP1840832A1 (en) | Pattern information registration apparatus, pattern information registration method, pattern information registration program and pattern matching system | |
US10546106B2 (en) | Biometric verification | |
US6961449B2 (en) | Method of correlation of images in biometric applications | |
US20070047771A1 (en) | Authentication apparatus | |
US7043083B2 (en) | Pattern-collating device, pattern-collating method and pattern-collating program | |
US11500975B2 (en) | Authentication device, authentication method, and computer program | |
JPH11144056A (en) | Electronic signature matching method and system therefor | |
US7155040B2 (en) | Generation of quality field information in the context of image processing | |
JP2006244365A (en) | Biometrics collation method | |
EP4167179A1 (en) | Authentication method, authentication program, and information processing device | |
KR20070019363A (en) | Method of biomass authentication for reducing FAR | |
US10956717B2 (en) | Collator and method for displaying result of collation | |
JPH11306352A (en) | Method and device for estimating authentication accuracy of somatological feature and recording medium | |
JPS634381A (en) | Fingerprint collating device | |
KR100416038B1 (en) | Electronic signature certification method using dynamic program and neural network |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AS | Assignment |
Owner name: SONY CORPORATION, JAPAN Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:ABE, HIROSHI;REEL/FRAME:018899/0467 Effective date: 20061113 |
|
FEPP | Fee payment procedure |
Free format text: PAYOR NUMBER ASSIGNED (ORIGINAL EVENT CODE: ASPN); ENTITY STATUS OF PATENT OWNER: LARGE ENTITY |
|
STCF | Information on status: patent grant |
Free format text: PATENTED CASE |
|
FPAY | Fee payment |
Year of fee payment: 4 |
|
MAFP | Maintenance fee payment |
Free format text: PAYMENT OF MAINTENANCE FEE, 8TH YEAR, LARGE ENTITY (ORIGINAL EVENT CODE: M1552); ENTITY STATUS OF PATENT OWNER: LARGE ENTITY Year of fee payment: 8 |
|
FEPP | Fee payment procedure |
Free format text: MAINTENANCE FEE REMINDER MAILED (ORIGINAL EVENT CODE: REM.); ENTITY STATUS OF PATENT OWNER: LARGE ENTITY |
|
LAPS | Lapse for failure to pay maintenance fees |
Free format text: PATENT EXPIRED FOR FAILURE TO PAY MAINTENANCE FEES (ORIGINAL EVENT CODE: EXP.); ENTITY STATUS OF PATENT OWNER: LARGE ENTITY |
|
STCH | Information on status: patent discontinuation |
Free format text: PATENT EXPIRED DUE TO NONPAYMENT OF MAINTENANCE FEES UNDER 37 CFR 1.362 |
|
FP | Lapsed due to failure to pay maintenance fee |
Effective date: 20230322 |